Exploring and Investigating Real-time Data Analytics and Visualization Solutions for Industrial IoT Use Cases
No Thumbnail Available
URL
Journal Title
Journal ISSN
Volume Title
Perustieteiden korkeakoulu |
Master's thesis
Authors
Date
2018-08-20
Department
Major/Subject
Digital Media Technology
Mcode
SCI3023
Degree programme
Master's Programme in ICT Innovation
Language
en
Pages
85+7
Series
Abstract
The master thesis is to develop an all-around real-time data visualization and analytics solution under the background of ABB. The solutions were built based on a ‘device-edge-cloud’ digital ecosystem to ‘digest’ a large amount of IIoT data and transform it into effective business actions. Data visualization and analytics act as a crucial linchpin for the transformation, and this progress should be completed in a timely fashion in order to preserve the maximum value of data for the final decision making. Within such a context, the topic is proposed and approved for execution. The research starts with investigating and clarifying the expectations on the final solutions from the angle of end-user so that the follow-up development work will be guided with a standard. Here, one fundamental standard for the solution development is to guarantee near real-time data delivery latency. So, the next chapter of the thesis introduces a real-time data processing architecture: Lambda Architecture, for supporting it. The final result is presented as four built solutions based on Lambda Architecture as well as a comprehensive performance analysis for each solution. And, the work initiates with four target products that have been used or stimulated the using interest of some business units in the company. And, they are respectively Power BI, Grafana, Time Series Insights and Splunk. In order to complete the rest of functionality of Lambda Architecture, other related products were selected and then mapped into each target product based solution. The performance analysis is conducted from four main aspects: architecture implementation possibility and difficulty, real-time data analytics capability, real-time visualization capability and finally other client-end application related aspects. And according to the final result, it can be concluded that except Time Series Insights based solution, the other three solutions could be selected depending on different scenarios. More specifically, Power BI based solution is the most comprehensive, Grafana based is the most cost-effective solution and Splunk based solution is the most convenient one. The end of research gives some further feedbacks to the most of current IIoT digital platforms, and it would expect that the digital platform would provide Data Analytics as a Service (DAaaS), instead of only Data as a Service (DaaS). For its current digital visualization and analytics services, it should be developed into a higher level of intelligence, which means that it can handle the content curation and decision making & action smartly and automatically.Description
Supervisor
Vuorimaa, PetriThesis advisor
Niinisto, AriKeywords
Industrial Internet of Things (IIoT), real-time data, stream data analytics, batch data analytics